Sigma receptors were initially proposed as opioid, and later phencyclidine receptors, and were finally demonstrated to represent unique binding sites in mammalian brain and peripheral tissues that are expressed throughout the CNS and have been implicated in a variety of physiological functions and disease states. Two subtypes of receptors have been distinguished molecularly and pharmacologically. Although the psychomotor stimulant and reinforcing effects of cocaine are primarily mediated through stimulation of dopamine (DA) neurotransmission by inhibiting dopamine reuptake, cocaine also shows a moderate affinity for Sigma receptors, and it has been suggested that some cocaine-induced effects could be mediated or modulated by its own actions at sigma 1 and/or sigma 2 receptor subtypes. In addition several studies have shown antagonism of various effects of cocaine by sigma receptor antagonists, and chronic exposure to cocaine has neuroplastic effects that are prevented by the sigma receptor antagonists. Though several effects of cocaine related to its abuse are antagonized by sigma receptor antagonists, a study from another laboratory showed that the standard &#963;R antagonist, BD1047, had no effects on cocaine self -administration, though it did modestly modify the reinstatement of cocaine self-administration. To ensure that the inactivity of BD1047 was not simply an idiosyncratic lack of effect of that particular &#963;R antagonist, we studied the effects of several &#963;R antagonists (BD1008, BD1063, AC927, NE100) on cocaine self-administration. None of the &#963;R antagonists at any dose significantly altered cocaine self-administration across the entire cocaine dose-effect curve. However, &#963;R agonists (DTG, PRE-084) shifted the cocaine doseeffect curve leftward. In potentiating the self-administration of cocaine PRE-084 was three times more potent than DTG. The potentiation of cocaine self-administration by &#963;R agonists was similar to the potentiation produced by standard DA uptake inhibitors, which are also self-administered. This finding suggested that &#963;R agonists would also be self-administered. In contrast to the lack of self-administration in nave subjects, the &#963;R agonists were self-administered when substituted for cocaine. PRE-084 was three times more potent than DTG, a relative potency similar to that for the potentiation of cocaine self-administration. This outcome was remarkable as none of the modest behavioral effects of &#963;R agonists resemble those of any known drug of abuse. For example, both PRE-084 and DTG were inactive in subjects trained to discriminate cocaine from saline injections, and produced no grossly observable stimulant-like or sedative effects. We further studied the conditions under which &#963;R agonists would be self-administered. Our first question was whether &#963;R agonists would be reinforcing in subjects without cocaine self-administration experience. Subjects were surgically prepared with venous catheters and allowed to self-administer PRE-084 (at the dose that maintained maximum self-administration in cocaine experienced subjects) for 28 sessions, more than sufficient for the acquisition of cocaine self-administration. In these cocaine-nave subjects there was no acquisition of PRE-084 self-administration, or self-administration of PRE-084 over a 100-fold dose range in further studies. Negative effects were also obtained with the &#963;R agonist, (+)-pentazocine. These subjects were then allowed to self-administer cocaine. Acquisition of cocaine self-administration was retarded compared to naive subjects, but developed and reached asymptote by the last three of 14 sessions. Once self-administration was maintained by cocaine, PRE-084 was introduced again in place of cocaine. Over the course of the next 10 sessions self administration was maintained by PRE-084 at asymptotic levels. Over the next seven sessions, PRE-084 could still be self-administered, but it was contingent on responses emitted on a previously inactive lever. In the first of these sessions, rates of responding decreased on the previously active lever and simultaneously increased on the lever that was made newly active. When saline was substituted for PRE-084, self-administration ceased over the next 10 sessions. PRE-084 was subsequently made available for responding on the left lever, and that response increased to the previous asymptotic levels. Effects were replicated with the &#963;R agonist, (+)-pentazocine. The dose-effect curve for PRE-084 self-administration now showed the characteristic bell-shape seen with other drugs of abuse, with maximal response rates at 0.32 mg/kg/injection. Thus cocaine exposure appeared to recruit &#963;Rs into reinforcement mechanisms, and the change appeared to be permanent. After establishment of PRE-084 self-administration the pharmacological mechanisms were assessed with specific receptor antagonists. The antagonism of PRE-084 self administration was compared to the effects of those same antagonists on cocaine self administration. The effects of DA receptor antagonists were characteristic of those in the literature. Standard D1, D2 and non-selective DA receptor antagonists (SCH39166, L-741,626, haloperidol, respectively) blocked the self-administration of cocaine, often exhibited as dose-dependent rightward shifts in the cocaine dose-effect curve. Consistent with previous results, the &#963;R antagonist, BD1063 had no effects on cocaine self-administration. In contrast to these effects on cocaine self-administration, neither D1 nor D2 DA antagonist, nor their combination, was effective in altering PRE-084 self-administration. However BD1063, which was ineffective against cocaine, dose-dependently blocked PRE-084 self-administration. An interesting outcome was obtained with haloperidol, which blocked both cocaine and PRE-084 self-administration. Haloperidols DA antagonist effects are well documented. In addition, haloperidol also possesses &#963;R antagonist effects, which likely contributed to the blockade of PRE-084 self-administration. These pharmacological studies clearly establish that once induced, &#963;R-agonist reinforcement is independent of DA mechanisms. The effects of PRE-084 on DA concentrations in the nucleus accumbens shell were also examined in cocaine experienced and naive subjects. PRE-084 produced a dose-related increase in DA concentrations after its intravenous injection. The significant increase in DA concentration was obtained at a dose of 10 mg/kg of PRE-084, a dose 100-fold higher than that effective in the self-administration procedure. Antagonism studies indicated that the effect of PRE-084 is not a &#963;R-mediated. Further there was no significant difference in the DA effects of PRE-084 in rats with and without cocaine self-administration experience indicating that DA is not involved in PRE-084 self-administration. Critical features involved in the cocaine experience that trigger reinforcing effects of &#963;R agonists are still under study. The necessity of a contingency relation between cocaine injection and a response was examined using the yoked control procedure. A master group was allowed to self administer cocaine (0.32 mg/kg/injection). Yoked-control groups were treated identically to the master group, but received injections of cocaine or saline injections coincident with the master group's self-administrations. Subjects from both yoked control groups pressed the lever infrequently, if at all. After acquisition of cocaine self-administration in the master group, only those subjects self-administered PRE-084. Thus it appears that a contingency relation between the response and the cocaine injection is a necessary component of the induction of &#963;R agonist self-administration.